Invalid geometries

[eric-g-97477]

I need to be able to compute a "center" point for the largest landmasses in the geometries for the countries. The shapely.ops.polylabel function implements a good algorithm for doing that.

Note that if I use the 50m (https://github.com/AshKyd/geojson-regions/blob/master/countries/50m/SDN.geojson) version of the file, polylabel works.

Unfortunately, the geometry defined by SDN.geojson is invalid. Shapely's explain_validity function returns the following information:

0    Self-intersection[33.9633927949711 9.464285229...
Name: geometry, dtype: object

make_valid generates the following error:

---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
Input In [5], in <cell line: 355>()
    353 gdf = geopandas.GeoDataFrame.from_features( features, crs = 4326 )   
    354 print( explain_validity( gdf.geometry ) )
--> 355 print( make_valid( gdf.geometry ) )
    356 gdf.plot( figsize = ( 20, 20 ) )
    357 center = polylabel( gdf.geometry.values[0], tolerance = .1 ) 

File shapely/validation.py:53, in make_valid(ob)
     29 def make_valid(ob):
     30     """
     31     Make the input geometry valid according to the GEOS MakeValid algorithm.
     32 
   (...)
     51 
     52     """
---> 53     if ob.is_valid:
     54         return ob
     55     return shapely.make_valid(ob)

File pandas/core/generic.py:1527, in NDFrame.__nonzero__(self)
   1525 @final
   1526 def __nonzero__(self) -> NoReturn:
-> 1527     raise ValueError(
   1528         f"The truth value of a {type(self).__name__} is ambiguous. "
   1529         "Use a.empty, a.bool(), a.item(), a.any() or a.all()."
   1530     )

ValueError: The truth value of a Series is ambiguous. Use a.empty, a.bool(), a.item(), a.any() or a.all().

I suggest updating how the geometries are computed so they are guaranteed to be valid. There are several other invalid geometries in the dataset.

A script that generates the error is:

The data is the SDN.geojson file located at https://github.com/AshKyd/geojson-regions/blob/master/countries/110m/SDN.geojson
(removed some of the "properties" for simplicity). It will draw correctly.

import json
from shapely.ops import polylabel
import geopandas
import os

data = {
  "type": "Feature",
  "properties": {
    "featurecla": "Admin-0 country",
    "filename": "SDN.geojson"
  },
  "geometry": {
    "type": "Polygon",
    "coordinates": [
      [
        [
          24.567369012152085,
          8.229187933785468
        ],
        [
          23.805813429466752,
          8.666318874542526
        ],
        [
          23.459012892355986,
          8.954285793488893
        ],
        [
          23.394779087017184,
          9.265067857292223
        ],
        [
          23.55724979014283,
          9.681218166538684
        ],
        [
          23.554304233502194,
          10.089255275915308
        ],
        [
          22.97754357269261,
          10.71446259199854
        ],
        [
          22.864165480244225,
          11.142395127807546
        ],
        [
          22.87622,
          11.384610000000123
        ],
        [
          22.50869,
          11.67936
        ],
        [
          22.49762,
          12.26024
        ],
        [
          22.28801,
          12.64605
        ],
        [
          21.93681,
          12.588180000000136
        ],
        [
          22.03759,
          12.95546
        ],
        [
          22.29658,
          13.37232
        ],
        [
          22.18329,
          13.78648
        ],
        [
          22.51202,
          14.09318
        ],
        [
          22.30351,
          14.32682
        ],
        [
          22.56795000000011,
          14.944290000000137
        ],
        [
          23.024590000000103,
          15.68072
        ],
        [
          23.886890000000108,
          15.61084
        ],
        [
          23.83766000000014,
          19.580470000000105
        ],
        [
          23.850000000000136,
          20
        ],
        [
          25.000000000000114,
          20.00304
        ],
        [
          25,
          22
        ],
        [
          29.02,
          22
        ],
        [
          32.9,
          22
        ],
        [
          36.86623,
          22
        ],
        [
          37.1887200000001,
          21.01885
        ],
        [
          36.96941,
          20.83744000000013
        ],
        [
          37.11470000000014,
          19.80796
        ],
        [
          37.4817900000001,
          18.61409
        ],
        [
          37.86276,
          18.36786
        ],
        [
          38.410089959473225,
          17.998307399970315
        ],
        [
          37.90400000000011,
          17.42754
        ],
        [
          37.16747,
          17.263140000000135
        ],
        [
          36.852530000000115,
          16.95655
        ],
        [
          36.75389,
          16.29186
        ],
        [
          36.32322,
          14.82249
        ],
        [
          36.42951,
          14.42211
        ],
        [
          36.27022,
          13.563330000000121
        ],
        [
          35.86363,
          12.57828
        ],
        [
          35.26049,
          12.08286
        ],
        [
          34.83163000000013,
          11.318960000000118
        ],
        [
          34.73115000000013,
          10.910170000000107
        ],
        [
          34.25745,
          10.63009
        ],
        [
          33.96162,
          9.58358
        ],
        [
          33.97498,
          8.68456
        ],
        [
          33.963392794971185,
          9.464285229420625
        ],
        [
          33.82496348090751,
          9.484060845715362
        ],
        [
          33.84213085302815,
          9.981914637215993
        ],
        [
          33.72195924818311,
          10.325262079630193
        ],
        [
          33.206938084561784,
          10.720111638406593
        ],
        [
          33.086766479716744,
          11.441141267476496
        ],
        [
          33.206938084561784,
          12.179338268667095
        ],
        [
          32.743419037302544,
          12.248007757149992
        ],
        [
          32.67474954881965,
          12.02483191958072
        ],
        [
          32.073891524594785,
          11.973329803218519
        ],
        [
          32.31423473428475,
          11.68148447716652
        ],
        [
          32.400071594888345,
          11.080626452941488
        ],
        [
          31.850715687025513,
          10.531270545078826
        ],
        [
          31.35286189552488,
          9.810240916008695
        ],
        [
          30.837840731903384,
          9.70723668328452
        ],
        [
          29.996639497988554,
          10.290927335388687
        ],
        [
          29.61895731133285,
          10.084918869940225
        ],
        [
          29.515953078608618,
          9.793073543888056
        ],
        [
          29.000931914987177,
          9.60423245056029
        ],
        [
          28.966597170745786,
          9.398223985111656
        ],
        [
          27.970889587744352,
          9.398223985111656
        ],
        [
          27.833550610778786,
          9.60423245056029
        ],
        [
          27.112520981708883,
          9.638567194801624
        ],
        [
          26.752006167173818,
          9.466893473594496
        ],
        [
          26.477328213242515,
          9.552730334198088
        ],
        [
          25.962307049621018,
          10.136420986302426
        ],
        [
          25.790633328413946,
          10.411098940233728
        ],
        [
          25.069603699343986,
          10.273759963267992
        ],
        [
          24.794925745412684,
          9.810240916008695
        ],
        [
          24.53741516360202,
          8.91753756573172
        ],
        [
          24.19406772118765,
          8.728696472403897
        ],
        [
          23.886979580860668,
          8.619729712933065
        ],
        [
          24.567369012152085,
          8.229187933785468
        ]
      ]
    ]
  }
}

features = {
    "type": "FeatureCollection",
    "features": [data]
}

gdf = geopandas.GeoDataFrame.from_features( features, crs = 4326 )   
print( explain_validity( gdf.geometry ) )
gdf.plot( figsize = ( 20, 20 ) )
center = polylabel( gdf.geometry.values[0], tolerance = .1 ) 
print( center )